The Bicameral Cache: a split cache for vector architectures

TL;DR

The paper proposes the Bicameral Cache, a split cache organization for vector architectures that segregates scalar and vector data to improve performance, with an added prefetching mechanism to enhance cache hit rates.

Contribution

It introduces a novel split cache design for vector architectures that isolates scalar and vector data, along with a prefetching mechanism to boost spatial locality and performance.

Findings

Speeds up stride-1 vector benchmarks

Minimal impact on non-stride-1 benchmarks

Prefetching consistently improves cache performance

Abstract

The Bicameral Cache is a cache organization proposal for a vector architecture that segregates data according to their access type, distinguishing scalar from vector references. Its aim is to avoid both types of references from interfering in each other's data locality, with a special focus on prioritizing the performance on vector references. The proposed system incorporates an additional, non-polluting prefetching mechanism to help populate the long vector cache lines in advance to increase the hit rate by further exploiting the spatial locality on vector data. Its evaluation was conducted on the Cavatools simulator, comparing the performance to a standard conventional cache, over different typical vector benchmarks for several vector lengths. The results proved the proposed cache speeds up performance on stride-1 vector benchmarks, while hardly impacting non-stride-1's. In addition, the prefetching feature consistently provided an additional value.